Elastic search language configuration - symfony

Question

I work on a project with symfony and sonata admin,
my project will be in two language: fr and de so i used elastic search so i installed "friendsofsymfony/elastica-bundle": "3.2.1".
in the elasticconfiguration file how can add the language configuration?
this is my fos elastic bundle configuration file:
fos_elastica:
clients:
default: { host: %elastic_host%, port: %elastic_port% }
indexes:
myproject:
settings:
index:
analysis:
analyzer:
custom_analyzer :
type : custom
tokenizer: nGram
filter : [stopwords, asciifolding ,lowercase, elision, worddelimiter]
custom_search_analyzer :
type : custom
tokenizer: standard
filter : [stopwords, asciifolding ,lowercase, elision, worddelimiter]
tokenizer:
nGram:
type: nGram
min_gram: 3
max_gram: 20
filter:
elision:
type: elision
articles: [l, m, t, qu, n, s, j, d]
stopwords:
type: stop
stopwords: [_french_]
ignore_case : true
worddelimiter:
type: word_delimiter
types:
produit:
mappings:
titre: ~
active:
type: boolean
descriptifTexteOriginal:
index_analyzer: custom_analyzer
search_analyzer : custom_search_analyzer
type: string
norms: { enabled: false }
explain: true
index_options: freqs
persistence:
driver: orm
model: myproject\BOBundle\Entity\Produit
finder: ~
provider: ~
listener: ~

Answer 1

In base, we will about two situtaion. First one is you have a field and it can store
two or more different language text. For example, you have a title field which can
store english or turkish. Not both of them is at the same time. The second situation
is, your data have multiple language value in one document.
In my opinion, there is two way of this problem, too. First is separate your fields
and second is separating index. Let's explain both of them to you:
Think about you have the following data for your products on e-commerce website:
id => identify of your data
name => name of the product
price => price of the product
Our first opinion has multiple option. Now, let's look our first option of the first
opinion and create a mapping:
PUT 44369578
{
"mappings": {
"product": {
"properties": {
"id": { "type": "integer" },
"name": {
"type": "text",
"fields": {
"en": {
"type": "text",
"analyzer": "englishAnalyzer"
},
"tr": {
"type": "text",
"analyzer": "turkishAnalyzer"
}
}
},
"price": { "type": "double"}
}
}
}
}
There is a name field which can take english or turkish value
And in this field only can handle one of them. Not both of them
When you try to improve your language support, you should change your mapping, too.
Then, let's look our second options of the first opinion.
PUT 44369578
{
"mappings": {
"product": {
"properties": {
"id": { "type": "integer" },
"name_en": {"type": "text", "analyzer": "englishAnalyzer" },
"name_tr": {"type": "text", "analyzer": "turkishAnalyzer" },
"price": { "type": "double"}
}
}
}
}
In this example there is two field for our two language. And we can
save english and turkish values for our documents at the same time.
When you try to improve your language support, you should change your mapping, too.
These are two options of the first opinion.
Now, lets look our second opinion. In this example, we have same data structure for
product documents. But we create multiple indexes for each languages.
PUT 44369578
{
"mappings": {
"product": {
"properties": {
"id": { "type": "integer" },
"name": {"type": "text", "analyzer": "englishAnalyzer" },
"price": { "type": "double"}
}
}
}
}
PUT 44369578
{
"mappings": {
"product": {
"properties": {
"id": { "type": "integer" },
"name": {"type": "text", "analyzer": "turkishAnalyzer" },
"price": { "type": "double"}
}
}
}
}
In this opinion, we have two documents for each document and we try to merge the
documents on our application while we are searching documents. This is odd part of
this opinion. But we can create a new index while we create a new language support.
Conclusion, As far as, I know these three methods. You can select one of them and
don't forget, each election will get some features but take away some features.